Material properties of the brain are critical to its biomechanical responses under quasi-static (e.g., neurosurgery) or dynamic impact (e.g., traumatic brain injury) conditions. However, they remain poorly understood despite their critical importance and more than fifty years of research on brain biomechanics. Consequently, the brain has been characterized with a number of models and methods that have led to variable results (e.g., shear modulus estimates differ by as much as two orders in magnitude or more), in part because of limited access to the in vivo brain. This proposal leverages our longstanding and ongoing research on brain shift compensation in image-guided neurosurgery to provide a subject pool undergoing open cranial surgeries where measurements of parenchymal surface deformation will be used to estimate material properties of the in vivo brain. The project has two specific aims. The first aim will generate a database of parenchymal surface force/displacement responses under quasi-static in vivo conditions. As part of this effort, we will develop an indentation technique and incorporate it with a stereovision motion tracking system in the operating room (OR). Based on our preliminary simulations of brain response to indentation and pilot results on cortical surface deformation collected with stereovision motion tracking, we expect that the data collected in proposed project will be an important contribution to the research community through which to improve our understanding of in vivo brain biomechanics. The second aim of the project will utilize this database to evaluate and estimate material property constants for representative material models. By comparing outputs from subject-specific finite element models that use indentation forces and material properties as input, the relative applicability of these simulations will be evaluated. We will then parametrically identify the most critical property constants in each material model in order to estimate the values which improve model prediction accuracy relative to the in vivo force/displacement observations. By completing the proposed studies, we expect to improve our understanding of how current models and material property estimates capture specific characteristics of brain biomechanics in vivo. This information will set the stage for further studies seeking to advance the brain models that are available and their validity in simulations of biomechanical processes and responses in the brain in vivo. We will exploit the collective expertise of a team of investigators in image-guided neurosurgery, traumatic brain injury, image analysis, and biomechanical computational modeling to meet the clinical and technical challenges presented in this proposal. ! !